Abstract
Air pollution (AP), intensified by industrialization and urbanization, is a key environmental factor linked to rheumatoid arthritis (RA). However, its molecular and immunological impact on RA remains unclear. This study integrates epidemiological data, bioinformatics, single-cell transcriptomics, and animal models to investigate how AP contributes to the development of RA. Global epidemiological analysis shows rising RA prevalence in over 95% of countries. Mendelian randomization analysis indicated a positive correlation between PM(10) exposure and the risk of RA. Machine learning identifies Formin Binding Protein 1 (FNBP1) as a key air pollution-related gene (APRG), with decreasing expression in RA patients and strong correlation with disease activity. PM(10) exposure may impair natural killer (NK) cell differentiation and cytotoxicity by suppressing FNBP1 expression, ultimately weakening immune surveillance and exacerbating inflammatory responses. Furthermore, by integrating single-cell sequencing, animal models, and human-derived cell experiments, we demonstrated that PM(10) exposure aggravates inflammation and joint damage in a collagen-induced arthritis (CIA) model. Mechanistically, PM(10) likely impairs the cytotoxic function of CD56(dim) NK cells through the modulation of FNBP1. Taken together, our research results have unveiled a completely novel mechanistic hypothesis regarding the onset and development of RA, the "PM(10)-FNBP1-NK cells" axis.